Reframing the 'Everything is everywhere'debate: evidence for high ...

60
Aquat Microb Ecol 41: 55–65, 2005
Table 3. Percent divergence within and between clades of Halteria grandinella
and its relatives, calculated using uncorrected distances
Clade a 1 2 3 4 5 6 7
1 0.13
2 2.93 0.25
3 1.95 3.71 0.25
4 6.69 7.31 6.49 0.39
5 6.05 7.64 4.88 8.22 0.20
6 5.07 5.88 4.10 7.22 5.23 0.41
7 5.61 7.41 5.20 7.80 6.15 4.55 0.20
a Descriptions of populations/clades refer to Table 1 and Fig. 1
Fig. 3. (A) Map indicating the locations of Strombidium oculatum and S.
stylifer sampling sites, from which at least 8 clones were sequenced (see
Table 2). Only a few clones were sequenced from many pools and although
they are included in the genealogy, these populations are not displayed on
the map. Colors correspond to the different clades in the S. oculatum genealogy.
(B) Genealogy based on maximum-likelihood analysis of ITS data (see
‘Materials and methods’). The genealogy is rooted with previously published
sequences of Laboea strobila. Bootstrap values as in Fig. 1
from nearby Chesterfield, Massachusetts
(Clade 1; Fig. 1). While other clades may
be geographically isolated (e.g. Clades 4
and 6 were only found in South America
and Clade 5 in Africa), substantially more
data are required to determine the extent
to which gene flow occurs or is limited
among locations.
To further assess morphospecies descriptions,
protargol staining was carried
out on representative Halteria populations
from Clades 2, 3, 4, 5 and 7. Consistent
with our genealogy, Clades 5 to 7 are morphologically
distinct at this more detailed
level of microscopy (Table 4). Intriguingly,
the Southern Hemisphere clades (4 to 7)
are present in morphospecies that have
spiny cysts, while H. grandinella from
Europe and North America (Clades 1 to 3)
make smooth cysts (Fig. 2C,D, Table 4).
Analysis of the second species cluster,
the tide pool morphospecies Strombidium
oculatum and S. stylifer, collected over
short temporal (tidal) and both small and
large spatial scales reveals similar patterns
of high genetic diversity, except that
considerable variation also exists within
populations of these taxa (Figs. 3 & 4,
Tables 2 & 5). These species can be distinguished
by the habit of S. oculatum to
encyst with a tidal periodicity, and by a
posterior thorn-like projection that is usually
present on S. stylifer. In S. stylifer, the
adoral zone extends further down the ventral
side than that of S. oculatum, and the
girdle kinety of S. oculatum is closer to the
mouth. Both species inhabit similar environments,
contain a prominent eyespot,
and are mixotrophs, harboring ‘slave’
chloroplasts derived from green macroalgae
(McManus et al. 2004).
Genealogical analyses of 336 clones
from Strombidium oculatum and S. stylifer
isolated from both sides of the North
Atlantic and from the western South
Atlantic (coastal Brazil) reveal not 2 but
10 distinct clades (>2% sequence divergence;
Fig. 3). As with Halteria spp., there
is evidence of both genetic diversity and
high levels of gene flow in S. oculatum
despite the seemingly restricted habitat of
this tide pool ciliate (Fig. 3, Table 2). For
example, individuals represented in Clade
II are found in both the Irish Sea and on
the East Coast of North America. To our